Commit Graph

23 Commits

Author SHA1 Message Date
Joerg Wunsch
7998b1245d Make the FDC (state machine) state an enum, as opposed to an int
abusing a bunch of #defines, for clarity and better debugging support.
2001-06-05 21:01:46 +00:00
Poul-Henning Kamp
8177437d85 Complete the bio/buf divorce for all code below devfs::strategy
Exceptions:
        Vinum untouched.  This means that it cannot be compiled.
        Greg Lehey is on the case.

        CCD not converted yet, casts to struct buf (still safe)

        atapi-cd casts to struct buf to examine B_PHYS
2000-04-15 05:54:02 +00:00
Joerg Wunsch
5f830ea2cd Make the Y-E Data PCMCIA floppy of the Toshiba Libretto work under
-current.  It doesn't work yet as stable as the 3.x/PAO version of the
driver does, however, i get occasional `FDC direction bit not set' and
other weird messages, but it basically works at least.

The old (defunct) #ifdef FDC_YE stuff has been eliminated completely
now, PCMCIA-FDC specific functions have been implemented differently
where needed.

Unfortunately, due to the fact that the traditional PeeCee FDC with
its funny non-contiguous register space (one register for WD1003
harddisk controllers is interleaved into the FDC register set), and
Peter's subsequent changes involving two different bus space handles
for normal FDCs, the changes required for the Y-E stuff are more
complex than i'd love them to be.  I've done my best to keep the logic
for normal FDCs intact.

Since the Y-E FDC seems to lose interrupts after a FDC reset
sometimes, i've also replaced the timeout logic in fd_turnoff() to
generate an artificial pseudo interrupt in case of a timeout while the
drive has still outstanding transfers waiting.  This avoids the total
starvation of the driver that could be observed with highly damaged
media under 3.x/PAO.  This part of the patch has been revied by bde
previously.

I've fixed a number of occasions where previous commits have been
missing the encapuslation of ISA DMA related functions inside
FDC_NODMA checks.

I've added one call to SET_BCDR() during preparation of the format
floppy operation.  Floppy formatting has been totally broken before in
3.x/PAO (garbage ID fields have been written to the medium, causing
`wrong cylinder' errors upon media reading).  This is just black
magic, i don't have the slightes idea _why_ this needs to be but just
copied over the hack that has been used by the PAO folks in the normal
read/write case anyway.

The entired device_busy() stuff seems to be pointless to me.  In any
case, i had to add device_unbusy() calls symmetrical to the
device_busy() calls, otherwise the PCMCIA floppy driver could never be
deactivated.  (As it used to be, it caused a `mark the device busier
and busier' situation.)  IMHO, all block device drivers should be
marked busy based on active buffers still waiting for the driver, so
the device_unbusy() calls should probably go to biodone().  Only one
other driver (whose name escapes me at the moment) uses device_busy()
calls at all, so i question the value of all this...

I think this entire `device busy' logic simply doesn't fit for PCMCIA
&al.  It cannot be the decision of some piece of kernel software to
declare a device `busy by now, you can't remove it', when the actual
physical power of removing it is the user pulling the card.  The
kernel simply has to cope with the removal, however busy the device
might have been by the time of the removal, period.  Perhaps a force
flag needs to be added?

Upon inserting the card a second time, i get:

WARNING: "fd" is usurping "fd"'s cdevsw[]
WARNING: "fd" is usurping "fd"'s bmaj

I suspect this is related to the XXX comment at the call to
cdevsw_add().  Does anybody know what the correct way is to cleanup
this?
2000-03-18 18:27:01 +00:00
Bruce Evans
66aeaaf280 Removed some more vestiges of ft.
Fixed some style bugs.
2000-01-09 17:13:35 +00:00
Peter Wemm
b6e5f28e51 Futher cleanup.. "device_print_prettyname(); printf()" -> device_printf()
It seems that the IDE system uses 0x3f6 for itself, which conflicts with
fdc's default 0x3f0-3f7 allocation range. Sigh.  Work around this.
Use bus_set_resource() rather than allocating specific areas, it makes
the code a little cleaner.

Based on work by:	dfr
2000-01-08 09:33:09 +00:00
Warner Losh
3b1782065f Merge most of FDC_YE into the mainline driver.
o Rename FDC_PCMCIA to FDC_NODMA to allow systems that don't have dma
  for floppies.
o Remove all but two FDC_YE ifdefs.  They aren't needed.
o Move defines for YE_DATAPORT to fdreg.h.

Not fixed:
o The pccard probe/attach.  However, motivated individuals can more
  easily add this now.

This is a merge of changes I've had in my tree for a long time.  These
fixes were tested on my VAIO with its normal floppy.  Please let me
know if I broke anything.

Prodded by: Peter Wemm <peter@freebsd.org>
2000-01-06 07:13:54 +00:00
Peter Wemm
37286586ff Patch up some of the evilness left over from the early newbus porting.
In particular:
 - Don't leave resources allocated in the probe routine.  Allocate them
   during probe and release them.  Probe's job is to identify devices only.
 - Don't abuse the ivars pointer.. (!).  Create real ivars and use the
   proper access system.  (the bus_read_ivar method)
 - Don't add the children until attach() has successfully grabbed the
   hardware, otherwise there are potential leaks if attach fails.
2000-01-05 16:31:27 +00:00
Doug Rabson
427ccf0071 Add code to support ISA PnP. 1999-11-11 08:48:40 +00:00
Peter Wemm
c3aac50f28 $Id$ -> $FreeBSD$ 1999-08-28 01:08:13 +00:00
Peter Wemm
6182fdbda8 Bring the 'new-bus' to the i386. This extensively changes the way the
i386 platform boots, it is no longer ISA-centric, and is fully dynamic.
Most old drivers compile and run without modification via 'compatability
shims' to enable a smoother transition.  eisa, isapnp and pccard* are
not yet using the new resource manager.  Once fully converted, all drivers
will be loadable, including PCI and ISA.

(Some other changes appear to have snuck in, including a port of Soren's
 ATA driver to the Alpha.  Soren, back this out if you need to.)

This is a checkpoint of work-in-progress, but is quite functional.

The bulk of the work was done over the last few years by Doug Rabson and
Garrett Wollman.

Approved by:	core
1999-04-16 21:22:55 +00:00
Bruce Evans
e93e63cb39 Fixed corruption of the fd buffer queue. Once upon a time, the active
buffer had to be left on the head of the queue for [bufq]disksort()
to sort against.  This isn't right for devices that can support multiple
active i/o's, and only the fd driver did it.  "Fixing" this in rev.1.36
of ufs_disksubr.c broke the fd driver in much the same way as rev.1.52
of <sys/buf.h> broke it (see rev.1.119).

Bug reported and fix tested by:	dt
1999-01-15 09:15:27 +00:00
Warner Losh
69acd21dfd Add support for the YE-Data external PCMCIA floppy driver. This
floppy is used on the toshiba Libretto line of subnotebook computers.
It differs from a normal floppy in that you must use PIO rather than
DMA to transfer the data.

To enable this, you must add options "FDC_YE" to your kernel.  I don't
have a machine that has a floppy and a pcmcia slot to test to make
sure that this doesn't impact normal floppy units, so I've left this as
an option.

I have ported this to -current and made an attempt to ensure that the
indentation conforms to style(9), aka the bruce filter.

Reviewed by:	nate, markm
Submitted by:	David Horwitt (dhorwitt@ucsd.edu)
1998-12-12 08:16:01 +00:00
Bruce Evans
5c1a1eae83 Fixed error handling:
- Call isa_dmadone() whenever necessary to stop DMA and/or free bounce
  buffers.  Undead DMA corrupted the malloc freelist fairly consistently
  in the following configuration: SLICE kernel, 2 floppy drives, no disk
  in fd0, disk in fd1.
- Don't call fdc_reset() from fd_timeout().  Doing so gave an "extra"
  interrupt which was usually misinterpreted as being for completion
  of the next FDC command; the interrupt for completion of the next
  FDC command was then usually misinterpreted...  There were further
  complications for interrupts latched by the soft-spl mechanism so
  that they were delivered after all the h/w interrupts went away.
  This caused at least wrong head settle delays and may be why the
  FreeBSD floppy driver seems to munch floppies more than most floppy
  drivers.  The reset was unnecessary anyway in cases that didn't have
  the bug described next, since is was repeated a little later for
  the IOTIMEDOUT state.  The state machine has complications to handle
  resets correctly, so just use it.
- Don't call retrier() from fd_timeout().  The IOTIMEDOUT state needs
  to be processed next, and it isn't valid to set to that state if
  retrier() has aborted the current transfer.  Doing so caused null
  pointer panics after the previous bug was fixed.

Improved error handling:
- If an i/o is aborted, arrange to reset in the state machine before
  doing the next i/o.  New fdc flag for this.  This fixes spurious
  warnings and lengthy busy-waiting for the next i/o.
- Split STARTRECAL into RESETCOMPLETE and STARTRECAL and only check
  for the results from reset if we actually reset.  This fixes spurious
  warnings for other paths to STARTRECAL.  [Oops, it may break reset
  handling for motor-off resets.]

Cleanups in fd_timeout():
- Renamed to fd_iotimeout() to make it clearer that it is only used
  for i/o.
- Don't handle the bp == 0 case.  This case can't happen for i/o.
- Don't check for controller-busy.  We know it must be.
- Don't print anything.  retrier() already prints too much for normal
  errors.
- Fudge the state differently so that the state machine advances
  fdc->retry and the status is invalid (perhaps this should fudge a
  valid state like the one for WP).
- Style fixes.
1998-07-29 13:00:42 +00:00
Bruce Evans
a838d83d8f Don't pretend to support ix86's with 7-bit ints by using longs just to
ensure 8-bit variables.  Doing so mainly bogotified some printf formats.
1998-07-11 06:35:39 +00:00
Tor Egge
d66c374f58 Enable the FIFO on enhanced floppy controllers. This reduces the
number of dma overruns/underruns for systems under heavy dma load.
As a side effect, broken enhanced floppy controllers that sometimes
don't detect dma overruns/underruns will give less errors.

Reviewed by:	j@uriah.heep.sax.de (J Wunsch)
1997-09-17 20:16:17 +00:00
Peter Wemm
6875d25465 Back out part 1 of the MCFH that changed $Id$ to $FreeBSD$. We are not
ready for it yet.
1997-02-22 09:48:43 +00:00
Jordan K. Hubbard
1130b656e5 Make the long-awaited change from $Id$ to $FreeBSD$
This will make a number of things easier in the future, as well as (finally!)
avoiding the Id-smashing problem which has plagued developers for so long.

Boy, I'm glad we're not using sup anymore.  This update would have been
insane otherwise.
1997-01-14 07:20:47 +00:00
Poul-Henning Kamp
17542807e5 Move from the old buf.b_actf to the new TAILQ(buf.b_act). 1996-05-03 14:57:27 +00:00
Joerg Wunsch
dc5df763e4 Peter's work to work around one of the most annoying bugs in the
floppy driver (or in the hardware?).  It turned out to be caused by
spurious interrupts, right after an FDC reset.

Also major cleanup in the low-level structure, there are now functions
performing error-checks for the FDC I/O.

Submitted by:	(mostly) Peter Dufault <dufault@FreeBSD.org>
1995-01-06 15:20:00 +00:00
Poul-Henning Kamp
702c623a8a Cosmetics. Silence gcc -Wall. Much more to do here :-( 1994-10-10 01:12:27 +00:00
David Greenman
3a2f7427fe Updated driver to the 1.1.5 version:
date: 1994/05/22 12:35:38;  author: joerg;  state: Exp;  lines: +6 -6
First round of floppy changes. Try making `fd' more robust.

New features:
  .  ioctl command for setting the drive type (density etc.); restricted
     to the super-user
  .  ioctl for getting/seting `drive options'; currently only option
     is FDOPT_NORETRY: inhibit the usual retries; used when verifying
     a newly formatted track

Fixes:
  .  function prototypes
  .  made all internal functions `static'
  .  cleaned up & corrected .h files
  .  restructured, to make the chaotic function sequence more rational
  .  compiled with -Wall, and cleared all warnings
  .  introduced a mirror for the (write-only) `digital output register',
     to avoid the current kludge
  .  device probing completed by seeking/recalibrating, and looking
     for track 0 being found
  .  holding the controller down in reset state while it is idle (and
     thus saving allot of headaches)
  .  make requests fail that are not a multiple of the (physical)
     sector size
  .  removed the fixed physical sector size (512 bytes), allowing for any
     size the controller could handle (128/256/512/1024 bytes)
  .  replaced some silly messages
  .  fixed the TRACE* macro usage, debugging reports should be complete
     now again (debugging output is HUGE! though)
  .  removed fd_timeout for SEEK command; seeks are always reported by
     the controller to succeed, since the `success' only refers to the
     controller's idea of success - there is no hardware line to tell about
     the seek end (other than the `track 0' line)
  .  catch SENSEI's that report about a `terminated due to READY changed'
     status - could happen after a controller reset
  .  converted ``hz / <something>'' divide operations to divisors that are
     powers of two, so gcc can optimize them into shifts
  .  write/format operations are checked against a write-protected medium
     now *prior* starting the operation
  .  error reports of `invalid command' and `wrong cylinder' will cause
     shortcuts in the retrier() now
  .  fixed a bug in the retrier() causing bogus block numbers to be reported
  .  fdformat() does care for errors now

Known Bugs:
  .  no attempts have been made (yet) to improve the performance
  .  sometimes, bogus ``seek/recalib failed'' messages are logged; this
     is still a bug in the driver, but it's not harmful since it's
     usually caught by the retrier()

Reviewed by:
Submitted by:
Obtained from:
1994-09-17 16:56:10 +00:00
Nate Williams
a60eff2770 From: Jim Babb <babb@sedhps01.mdc.com>
Date: Mon, 14 Feb 94 15:57:14 CST

This adds a copyright to the fdc.h file and fixes a bug in re-tries
during writes on a heavily loaded system.
1994-02-14 22:24:28 +00:00
Andrew Moore
9c2638e6dc Tentative fdc patches... 1994-02-07 22:12:45 +00:00